Unverified50% confidenceDecision RuleExact time
Refusal and handoff replies use fixed strings rather than model generation, to reduce the attack surface for prompt injection.
1
Sources
50%
Confidence
Long-term
Relevance
10/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe model that writes the answer should not decide whether to answer; every message first goes through a separate cheap classification call that returns one of three labels: in-scope answer, out-of-scope refusal, or handoff to a human.69% similarUnverifiedThe writable field set mechanism separately declares which fields are writable, and writes beyond the declared set should be rejected outright rather than truncated.68% similarUnverifiedPrompt injection is difficult to defend against with traditional firewalls because it does not rely on code vulnerabilities, but exploits the model's tendency to obey natural-language instructions, and its attack surface completely overlaps with normal functional input.67% similarUnverifiedPrompt injection attacks exploit the fundamental limitation of large language models being unable to strictly distinguish between instructions and data.66% similarUnverifiedAnswer masking refers to replacing the target field in the input text with a placeholder to prevent the model from directly copying it.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1034982API
curl https://kongchang.com/api/v1/knowledge/claims/1034982MCP
get_claim(id=1034982)